Legendary sports writer Red Smith once said that writing a column was simple: “Just sit down at a typewriter and bleed.” However, in 2026, blood will no longer be required. All you need to do is sit down at your laptop and ask Claude or ChatGPT to write the story for you.
It seems that such a conclusion can be drawn from a collection of reports from the journalistic front. Last month, my colleague Maxwell Zeff wrote about writers who unscrupulously create at least some of their prose through unregistered AI collaborators. The star of his piece was Alex Heath, a tech reporter who said he often commissions AI to write drafts based on his notes, interview transcripts and emails. The Wall Street Journal profile was published the same week. Fortuna reporter Nick Lichtenberg, who explained to the newspaper that he relies heavily on artificial intelligence in his work. He has written 600 stories since July; one day last February it had seven signatures.
Since reading these reports – fortunately written by human hands – I have had trouble sleeping. Until recently, the consensus was that using huge language models to actually create commercial prose was banned. Many publications, including WIREDhave strict guidelines for AI-generated text. We also don’t exploit it for editing, a less disturbing but still bothersome practice of several others cited in Zeff’s column. The book publishing world, trying to protect itself from the avalanche of self-published nonsense, continues to guard its catalogue; Most recently, Hachette Book Group withdrawn a novel that apparently relied too heavily on the LLM’s output. But as models turn out to be prose that is increasingly complex to distinguish from human work, the convenience and savings of using artificial intelligence for the complex work of writing threaten to seep into the mainstream. The walls start to crumble.
As you might expect, many people were unhappy to read about this development, especially those like me whose keyboards are dripping with blood. But the heroes of the story persist. It’s as if they felt the future was on their side. When I contacted Heath – whose work I respect – he confirmed that he had been rejected but shrugged it off. “I see artificial intelligence as a tool,” he says. “I don’t see it as replacing anything. The only thing that gets replaced is drudgery, which I didn’t want to do anyway.”
Of course, difficult writing is, for people like me, a critical aspect of the overall effort of engaging in the task of communicating effectively and clearly. Heath believes in connecting with readers through his writing – he claims to have trained his AI to sound like him, and his Substack contains personally written tidbits about what he does. On the other hand, he tells me that since talking to Zeff, he has almost “one-shotted” several of his columns. “When I say once, I mean I almost didn’t have to do anything,” he says. Heath, however, disputes the notion that allowing artificial intelligence to write prose for him means he bypasses the thought process that many believe can only occur through actual writing. “I just get rid of this very messy, painful, blank page from zero to one,” he says.
The Fortune writer who was the subject of the Journal article also faced consequences, not only from the public but also from his friends and colleagues. “I feel tension in my close and personal relationships.” Lichtenberg admitted in an interview with the Reuters Institute for Journalism Studies. In an email, Fortune editor-in-chief Alyson Shontell tried to dissuade me from the notion that artificial intelligence would take over reporting duties under her watch. “What’s important is [Lichtenberg] he doesn’t use it as a substitute for writing,” she wrote. “His stories are written artificially, not artificially. Still, the multitude of ambitious reports, analyzes and reworkings he produces are highly original.”